Hi, i am trying to use "mix" package for unrestricted general location model and multi-imputation on a dataset. While I am quite confident that data are ready, I receive errors on prelim.mix() step. My dataset is actually a data matrix with 365 obs and 11 vars. 10 vars are qualitative (ex-factor recoded as requested in ?prelim.mix) and ordered as needed in the first columns. NA status is good with no more than 1% per var. This is the output i receive when I run prelim.mix(mydata,10) Steps of EM: 1...2...3...Errore: NA/NaN/Inf in chiamata a funzione esterna (arg 6) Since the message is not so helpfull, could someone help me to debug? Like stlouis dataset, mine is a matrix and missing values are NA.